def generate_votes(self, input_data, by_name=True,
missing_strategy=LAST_PREDICTION,
add_median=False, add_min=False, add_max=False,
add_unused_fields=False):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict( \
input_data, by_name=by_name,
add_confidence=True,
add_distribution=True,
add_count=True,
add_median=add_median,
add_min=add_min,
add_max=add_max,
add_unused_fields=add_unused_fields,
missing_strategy=missing_strategy)
if model.boosting is not None:
votes.boosting = True
prediction_info.update( \
{"weight": model.boosting.get("weight")})
if model.boosting.get("objective_class") is not None:
prediction_info.update( \
{"class": model.boosting.get("objective_class")})
votes.append(prediction_info)
return votes
def generate_votes(self, input_data, by_name=True,
missing_strategy=LAST_PREDICTION,
add_median=False, add_min=False, add_max=False,
add_unused_fields=False):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict( \
input_data, by_name=by_name,
add_confidence=True,
add_distribution=True,
add_count=True,
add_median=add_median,
add_min=add_min,
add_max=add_max,
add_unused_fields=add_unused_fields,
missing_strategy=missing_strategy)
if model.boosting is not None:
votes.boosting = True
prediction_info.update( \
{"weight": model.boosting.get("weight")})
if model.boosting.get("objective_class") is not None:
prediction_info.update( \
{"class": model.boosting.get("objective_class")})
votes.append(prediction_info)
return votes
def predict(self, input_data, by_name=True, method=PLURALITY_CODE):
"""Makes a prediction based on the prediction made by every model.
:param input_data: Test data to be used as input
:param by_name: Boolean that is set to True if field_names (as
alternative to field ids) are used in the
input_data dict
:param method: numeric key code for the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
"""
# When only one group of models is found you use the
# corresponding multimodel to predict
input_data_array = self.format_input_data(input_data, by_name=by_name)
votes_split = []
options = None
for fun in self.predict_functions:
votes_split.append(fun(*input_data_array))
votes = MultiVote(votes_split, boosting_offsets=self.boosting_offsets)
if self.boosting is not None and not self.regression:
categories = [ \
d[0] for d in
self.fields[self.objective_id]["summary"]["categories"]]
options = {"categories": categories}
result = votes.combine(method=method, options=options)
return result
def predict(self, input_data, by_name=True, method=PLURALITY_CODE,
with_confidence=False, options=None):
"""Makes a prediction based on the prediction made by every model.
The method parameter is a numeric key to the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
3 - threshold filtered vote / doesn't apply:
THRESHOLD_CODE
"""
if len(self.models_splits) > 1:
# If there's more than one chunck of models, they must be
# sequentially used to generate the votes for the prediction
votes = MultiVote([])
for models_split in self.models_splits:
models = [retrieve_resource(self.api, model_id,
query_string=ONLY_MODEL)
for model_id in models_split]
multi_model = MultiModel(models, api=self.api)
votes_split = multi_model.generate_votes(input_data,
by_name=by_name)
votes.extend(votes_split.predictions)
else:
# When only one group of models is found you use the
# corresponding multimodel to predict
votes_split = self.multi_model.generate_votes(input_data,
by_name=by_name)
votes = MultiVote(votes_split.predictions)
return votes.combine(method=method, with_confidence=with_confidence,
options=options)
def aggregate_multivote(multivote,
options,
labels,
models_per_label,
ordered,
models_order,
label_separator=None):
"""Aggregate the model's predictions for multi-label fields in a
concatenated format into a final prediction
"""
if label_separator is None:
label_separator = ","
predictions = multivote.predictions
if ordered and models_per_label == 1:
# as multi-labeled models are created from end to start votes
# must be reversed to match
predictions.reverse()
else:
predictions = [
prediction
for (_, prediction) in sorted(zip(models_order, predictions),
key=lambda x: x[0])
]
if (labels is None or len(labels) * models_per_label != len(predictions)):
sys.exit("Failed to make a multi-label prediction. No"
" valid label info is found.")
prediction_list = []
confidence_list = []
# In the following case, we must vote each label using the models
# in the ensemble and the chosen method
if models_per_label > 1:
label_predictions = [
predictions[i:i + models_per_label]
for i in range(0, len(predictions), models_per_label)
]
predictions = []
for label_prediction in label_predictions:
label_multivote = MultiVote(label_prediction)
prediction_info = label_multivote.combine(method=AGGREGATION,
full=True,
options=options)
predictions.append({
'prediction': prediction_info["prediction"],
'confidence': prediction_info["confidence"]
})
for vote_index, vote_prediction in enumerate(predictions):
if ast.literal_eval(vote_prediction['prediction']):
prediction_list.append(labels[vote_index])
confidence = str(vote_prediction['confidence'])
confidence_list.append(confidence)
prediction = [
label_separator.join(prediction_list),
label_separator.join(confidence_list)
]
return prediction
def _generate_votes(self,
input_data,
missing_strategy=LAST_PREDICTION,
unused_fields=None):
""" Generates a MultiVote object that contains the predictions
made by each of the models. Please note that this function
calls a _predict method which assumes input data has been
properly checked against the model fields. Only casting
to the correct type will be applied.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model._predict( \
input_data,
missing_strategy=missing_strategy, unused_fields=unused_fields)
if model.boosting is not None:
votes.boosting = True
prediction_info.update( \
{"weight": model.boosting.get("weight")})
if model.boosting.get("objective_class") is not None:
prediction_info.update( \
{"class": model.boosting.get("objective_class")})
votes.append(prediction_info)
return votes
def predict(self, input_data, method=PLURALITY_CODE, full=False):
"""Makes a prediction based on the prediction made by every model.
:param input_data: Test data to be used as input
:param method: numeric key code for the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
"""
# When only one group of models is found you use the
# corresponding multimodel to predict
votes_split = []
options = None
count = 1
for fun in self.predict_functions:
prediction = fun(input_data)
prediction.update({"order": count, "count": 1})
count += 1
votes_split.append(prediction)
votes = MultiVote(votes_split, boosting_offsets=self.boosting_offsets)
if self.boosting is not None and not self.regression:
categories = [ \
d[0] for d in
self.fields[self.objective_id]["summary"]["categories"]]
options = {"categories": categories}
result = votes.combine(method=method, options=options, full=full)
if isinstance(result, dict):
del result['count']
return result
def generate_votes(self, input_data, by_name=True):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict(input_data, by_name=by_name, with_confidence=True)
prediction, confidence, distribution, instances = prediction_info
prediction_row = [prediction, confidence, order, distribution, instances]
votes.append_row(prediction_row)
return votes
def generate_probability_votes(self,
input_data,
by_name=True,
missing_strategy=LAST_PREDICTION,
method=PROBABILITY_CODE):
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
model.class_names = self.class_names
votes.probabilities = True
try:
if method == PROBABILITY_CODE:
prediction_info = model.predict_probability(
input_data,
by_name=by_name,
compact=True,
missing_strategy=missing_strategy)
elif method == CONFIDENCE_CODE:
prediction_info = model.predict_confidence(
input_data,
by_name=by_name,
compact=True,
missing_strategy=missing_strategy)
elif method == PLURALITY_CODE:
prediction_info = [0.0] * len(self.class_names)
prediction = model.predict(
input_data,
by_name=by_name,
missing_strategy=missing_strategy)
prediction_info[self.class_names.index(prediction)] = 1.0
else:
raise ValueError('%d is not a valid "method"' % method)
except (AttributeError, TypeError):
if method == PLURALITY_CODE:
prediction_info = [0.0] * len(self.class_names)
prediction = model.predict(input_data, by_name=by_name)
prediction_info[self.class_names.index(prediction)] = 1.0
else:
prediction_info = model.predict_probability(
input_data,
by_name=by_name,
compact=True)
votes.append(prediction_info)
return votes
def aggregate_multivote(multivote, options, labels, models_per_label, ordered,
models_order, label_separator=None):
"""Aggregate the model's predictions for multi-label fields in a
concatenated format into a final prediction
"""
if label_separator is None:
label_separator = ","
predictions = multivote.predictions
if ordered and models_per_label == 1:
# as multi-labeled models are created from end to start votes
# must be reversed to match
predictions.reverse()
else:
predictions = [prediction for (_, prediction)
in sorted(zip(models_order, predictions),
key=lambda x: x[0])]
if (labels is None or
len(labels) * models_per_label != len(predictions)):
sys.exit("Failed to make a multi-label prediction. No"
" valid label info is found.")
prediction_list = []
confidence_list = []
# In the following case, we must vote each label using the models
# in the ensemble and the chosen method
if models_per_label > 1:
label_predictions = [predictions[i: i + models_per_label] for
i in range(0, len(predictions),
models_per_label)]
predictions = []
for label_prediction in label_predictions:
label_multivote = MultiVote(label_prediction)
prediction, confidence = label_multivote.combine(
method=AGGREGATION, with_confidence=True, options=options)
predictions.append({'prediction': prediction,
'confidence': confidence})
for vote_index in range(0, len(predictions)):
if ast.literal_eval(predictions[vote_index]['prediction']):
prediction_list.append(labels[vote_index])
confidence = str(predictions[vote_index]['confidence'])
confidence_list.append(confidence)
prediction = [label_separator.join(prediction_list),
label_separator.join(confidence_list)]
return prediction
def i_create_a_multivote(step, predictions_file):
predictions_file = res_filename(predictions_file)
try:
with open(predictions_file, 'r') as predictions_file:
world.multivote = MultiVote(json.load(predictions_file))
except IOError:
assert False, "Failed to read %s" % predictions_file
def generate_votes(self, input_data, by_name=True,
missing_strategy=LAST_PREDICTION):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict(input_data, by_name=by_name,
with_confidence=True,
missing_strategy=missing_strategy)
prediction, confidence, distribution, instances = prediction_info
prediction_row = [prediction, confidence, order,
distribution, instances]
votes.append_row(prediction_row)
return votes
def generate_votes(self, input_data, by_name=True,
missing_strategy=LAST_PREDICTION,
add_median=False):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict(input_data, by_name=by_name,
add_confidence=True,
add_distribution=True,
add_count=True,
add_median=add_median,
missing_strategy=missing_strategy)
votes.append(prediction_info)
return votes
def predict(self, input_data, by_name=True, method=PLURALITY_CODE,
with_confidence=False, options=None):
"""Makes a prediction based on the prediction made by every model.
The method parameter is a numeric key to the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
3 - threshold filtered vote / doesn't apply:
THRESHOLD_CODE
"""
if len(self.models_splits) > 1:
# If there's more than one chunck of models, they must be
# sequentially used to generate the votes for the prediction
votes = MultiVote([])
for models_split in self.models_splits:
models = [retrieve_resource(self.api, model_id,
query_string=ONLY_MODEL)
for model_id in models_split]
multi_model = MultiModel(models, api=self.api)
votes_split = multi_model.generate_votes(input_data,
by_name=by_name)
votes.extend(votes_split.predictions)
else:
# When only one group of models is found you use the
# corresponding multimodel to predict
votes_split = self.multi_model.generate_votes(input_data,
by_name=by_name)
votes = MultiVote(votes_split.predictions)
return votes.combine(method=method, with_confidence=with_confidence,
options=options)
def generate_votes(self,
input_data,
by_name=True,
missing_strategy=LAST_PREDICTION,
add_median=False):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict(input_data,
by_name=by_name,
add_confidence=True,
add_distribution=True,
add_count=True,
add_median=add_median,
missing_strategy=missing_strategy)
votes.append(prediction_info)
return votes
def predict(self, input_data, by_name=True, method=PLURALITY_CODE,
with_confidence=False):
"""Makes a prediction based on the prediction made by every model.
The method parameter is a numeric key to the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
"""
votes = MultiVote([])
for models_split in self.models_splits:
models = [retrieve_model(self.api, model_id)
for model_id in models_split]
multi_model = MultiModel(models)
votes_split = multi_model.generate_votes(input_data,
by_name=by_name)
votes.extend(votes_split.predictions)
return votes.combine(method=method, with_confidence=with_confidence)
def generate_votes(self, input_data, missing_strategy=LAST_PREDICTION):
""" Generates a MultiVote object that contains the predictions
made by each of the models.
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict( \
input_data, missing_strategy=missing_strategy, full=True)
if model.boosting is not None:
votes.boosting = True
prediction_info.update( \
{"weight": model.boosting.get("weight")})
if model.boosting.get("objective_class") is not None:
prediction_info.update( \
{"class": model.boosting.get("objective_class")})
votes.append(prediction_info)
return votes
def predict(self, input_data, by_name=True, method=PLURALITY_CODE):
"""Makes a prediction based on the prediction made by every model.
The method parameter is a numeric key to the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
"""
votes = MultiVote([])
for order in range(0, len(self.models)):
model = self.models[order]
prediction_info = model.predict(input_data, by_name=by_name,
with_confidence=True)
prediction, confidence, distribution, instances = prediction_info
prediction_row = [prediction, confidence, order,
distribution, instances]
votes.append_row(prediction_row)
return votes.combine(method=method)
def predict(self, input_data, by_name=True, method=PLURALITY_CODE):
"""Makes a prediction based on the prediction made by every model.
:param input_data: Test data to be used as input
:param by_name: Boolean that is set to True if field_names (as
alternative to field ids) are used in the
input_data dict
:param method: numeric key code for the following combination
methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
"""
# When only one group of models is found you use the
# corresponding multimodel to predict
input_data_array = self.format_input_data(input_data, by_name=by_name)
votes_split = []
options = None
for fun in self.predict_functions:
votes_split.append(fun(*input_data_array))
votes = MultiVote(votes_split,
boosting_offsets=self.boosting_offsets)
if self.boosting is not None and not self.regression:
categories = [ \
d[0] for d in
self.fields[self.objective_id]["summary"]["categories"]]
options = {"categories": categories}
result = votes.combine(method=method, options=options)
return result
def read_votes(votes_files, to_prediction, data_locale=None):
"""Reads the votes found in the votes' files.
Returns a list of MultiVote objects containing the list of predictions.
votes_files parameter should contain the path to the files where votes
are stored
In to_prediction parameter we expect the method of a local model object
that casts the string prediction values read from the file to their
real type. For instance
>>> local_model = Model(model)
>>> prediction = local_model.to_prediction("1")
>>> isinstance(prediction, int)
True
>>> read_votes(["my_predictions_file"], local_model.to_prediction)
data_locale should contain the string identification for the locale
used in numeric formatting.
"""
votes = []
for order in range(0, len(votes_files)):
votes_file = votes_files[order]
index = 0
with UnicodeReader(votes_file) as rdr:
for row in rdr:
prediction = to_prediction(row[0], data_locale=data_locale)
if index > (len(votes) - 1):
votes.append(MultiVote([]))
distribution = None
instances = None
if len(row) > 2:
distribution = ast.literal_eval(row[2])
instances = int(row[3])
try:
confidence = float(row[1])
except ValueError:
confidence = 0.0
prediction_row = [
prediction, confidence, order, distribution, instances
]
votes[index].append_row(prediction_row)
index += 1
return votes
def batch_predict(self,
input_data_list,
output_file_path=None,
by_name=True,
reuse=False,
missing_strategy=LAST_PREDICTION,
headers=None,
to_file=True,
use_median=False):
"""Makes predictions for a list of input data.
When the to_file argument is set to True, the predictions
generated for each model are stored in an output
file. The name of the file will use the following syntax:
model_[id of the model]__predictions.csv
For instance, when using model/50c0de043b563519830001c2 to predict,
the output file name will be
model_50c0de043b563519830001c2__predictions.csv
On the contrary, if it is False, the function returns a list
of MultiVote objects with the model's predictions.
"""
add_headers = (isinstance(input_data_list[0], list)
and headers is not None
and len(headers) == len(input_data_list[0]))
if not add_headers and not isinstance(input_data_list[0], dict):
raise ValueError("Input data list is not a dictionary or the"
" headers and input data information are not"
" consistent.")
order = 0
if not to_file:
votes = []
for model in self.models:
order += 1
out = None
if to_file:
output_file = get_predictions_file_name(
model.resource_id, output_file_path)
if reuse:
try:
predictions_file = open(output_file)
predictions_file.close()
continue
except IOError:
pass
try:
out = UnicodeWriter(output_file)
except IOError:
raise Exception("Cannot find %s directory." %
output_file_path)
if out:
out.open_writer()
for index, input_data in enumerate(input_data_list):
if add_headers:
input_data = dict(zip(headers, input_data))
prediction = model.predict(input_data,
by_name=by_name,
with_confidence=True,
missing_strategy=missing_strategy)
if use_median and model.tree.regression:
# if median is to be used, we just place it as prediction
# starting the list
prediction[0] = prediction[-1]
prediction = prediction[:-1]
if to_file:
out.writerow(prediction)
else:
# prediction is a row that contains prediction, confidence,
# distribution, instances
prediction_row = prediction[0:2]
prediction_row.append(order)
prediction_row.extend(prediction[2:])
if len(votes) <= index:
votes.append(MultiVote([]))
votes[index].append_row(prediction_row)
if out:
out.close_writer()
if not to_file:
return votes
if (labels is None or
len(labels) * models_per_label != len(predictions)):
sys.exit("Failed to make a multi-label prediction. No"
" valid label info is found.")
prediction_list = []
confidence_list = []
# In the following case, we must vote each label using the models
# in the ensemble and the chosen method
if models_per_label > 1:
label_predictions = [predictions[i: i + models_per_label] for
i in range(0, len(predictions),
models_per_label)]
predictions = []
for label_prediction in label_predictions:
label_multivote = MultiVote(label_prediction)
prediction, confidence = label_multivote.combine(
method=method, with_confidence=True, options=options)
predictions.append({'prediction': prediction,
'confidence': confidence})
for vote_index in range(0, len(predictions)):
if ast.literal_eval(predictions[vote_index]['prediction']):
prediction_list.append(labels[vote_index])
confidence = str(predictions[vote_index]['confidence'])
confidence_list.append(confidence)
prediction = [label_separator.join(prediction_list),
label_separator.join(confidence_list)]
elif method == COMBINATION:
predictions = multivote.predictions
global_distribution = []
for prediction in predictions:
def batch_predict(self,
input_data_list,
output_file_path=None,
by_name=True,
reuse=False,
missing_strategy=LAST_PREDICTION,
headers=None,
to_file=True):
"""Makes predictions for a list of input data.
When the to_file argument is set to True, the predictions
generated for each model are stored in an output
file. The name of the file will use the following syntax:
model_[id of the model]__predictions.csv
For instance, when using model/50c0de043b563519830001c2 to predict,
the output file name will be
model_50c0de043b563519830001c2__predictions.csv
On the contrary, if it is False, the function returns a list
of MultiVote objects with the model's predictions.
"""
add_headers = (isinstance(input_data_list[0], list)
and headers is not None
and len(headers) == len(input_data_list[0]))
if not add_headers and not isinstance(input_data_list[0], dict):
raise ValueError("Input data list is not a dictionary or the"
" headers and input data information are not"
" consistent.")
order = 0
if not to_file:
votes = []
for model in self.models:
order += 1
if to_file:
output_file = get_predictions_file_name(
model.resource_id, output_file_path)
if reuse:
try:
predictions_file = open(output_file)
predictions_file.close()
continue
except IOError:
pass
try:
predictions_file = csv.writer(open(output_file, 'w', 0),
lineterminator="\n")
except IOError:
raise Exception("Cannot find %s directory." %
output_file_path)
for index, input_data in enumerate(input_data_list):
if add_headers:
input_data = dict(zip(headers, input_data))
prediction = model.predict(input_data,
by_name=by_name,
with_confidence=True,
missing_strategy=missing_strategy)
if to_file:
if isinstance(prediction[0], basestring):
prediction[0] = prediction[0].encode("utf-8")
predictions_file.writerow(prediction)
else:
prediction, confidence, distribution, instances = prediction
prediction_row = [
prediction, confidence, order, distribution, instances
]
if len(votes) <= index:
votes.append(MultiVote([]))
votes[index].append_row(prediction_row)
if not to_file:
return votes
or len(labels) * models_per_label != len(predictions)):
sys.exit("Failed to make a multi-label prediction. No"
" valid label info is found.")
prediction_list = []
confidence_list = []
# In the following case, we must vote each label using the models
# in the ensemble and the chosen method
if models_per_label > 1:
label_predictions = [
predictions[i:i + models_per_label]
for i in range(0, len(predictions), models_per_label)
]
predictions = []
for label_prediction in label_predictions:
label_multivote = MultiVote(label_prediction)
prediction, confidence = label_multivote.combine(
method=method, with_confidence=True, options=options)
predictions.append({
'prediction': prediction,
'confidence': confidence
})
for vote_index in range(0, len(predictions)):
if ast.literal_eval(predictions[vote_index]['prediction']):
prediction_list.append(labels[vote_index])
confidence = str(predictions[vote_index]['confidence'])
confidence_list.append(confidence)
prediction = [
label_separator.join(prediction_list),
label_separator.join(confidence_list)
]
def predict(self,
input_data,
method=None,
options=None,
missing_strategy=LAST_PREDICTION,
operating_point=None,
operating_kind=None,
median=False,
full=False):
"""Makes a prediction based on the prediction made by every model.
:param input_data: Test data to be used as input
:param method: **deprecated**. Please check the `operating_kind`
attribute. Numeric key code for the following
combination methods in classifications/regressions:
0 - majority vote (plurality)/ average: PLURALITY_CODE
1 - confidence weighted majority vote / error weighted:
CONFIDENCE_CODE
2 - probability weighted majority vote / average:
PROBABILITY_CODE
3 - threshold filtered vote / doesn't apply:
THRESHOLD_CODE
:param options: Options to be used in threshold filtered votes.
:param missing_strategy: numeric key for the individual model's
prediction method. See the model predict
method.
:param operating_point: In classification models, this is the point of
the ROC curve where the model will be used at.
The operating point can be defined in terms of:
- the positive_class, the class that is
important to predict accurately
- its kind: probability, confidence or voting
- its threshold: the minimum established
for the positive_class to be predicted.
The operating_point is then defined as a
map with three attributes, e.g.:
{"positive_class": "Iris-setosa",
"kind": "probability",
"threshold": 0.5}
:param operating_kind: "probability", "confidence" or "votes". Sets the
property that decides the prediction.
Used only if no operating_point is used
:param median: Uses the median of each individual model's predicted
node as individual prediction for the specified
combination method.
:param full: Boolean that controls whether to include the prediction's
attributes. By default, only the prediction is produced.
If set to True, the rest of available information is
added in a dictionary format. The dictionary keys can be:
- prediction: the prediction value
- confidence: prediction's confidence
- probability: prediction's probability
- path: rules that lead to the prediction
- count: number of training instances supporting the
prediction
- next: field to check in the next split
- min: minim value of the training instances in the
predicted node
- max: maximum value of the training instances in the
predicted node
- median: median of the values of the training instances
in the predicted node
- unused_fields: list of fields in the input data that
are not being used in the model
"""
# Checks and cleans input_data leaving the fields used in the model
new_data = self.filter_input_data( \
input_data,
add_unused_fields=full)
unused_fields = None
if full:
input_data, unused_fields = new_data
else:
input_data = new_data
# Strips affixes for numeric values and casts to the final field type
cast(input_data, self.fields)
if median and method is None:
# predictions with median are only available with old combiners
method = PLURALITY_CODE
if method is None and operating_point is None and \
operating_kind is None and not median:
# operating_point has precedence over operating_kind. If no
# combiner is set, default operating kind is "probability"
operating_kind = "probability"
if operating_point:
if self.regression:
raise ValueError("The operating_point argument can only be"
" used in classifications.")
prediction = self.predict_operating( \
input_data,
missing_strategy=missing_strategy,
operating_point=operating_point)
if full:
return prediction
return prediction["prediction"]
if operating_kind:
if self.regression:
# for regressions, operating_kind defaults to the old
# combiners
method = 1 if operating_kind == "confidence" else 0
return self.predict( \
input_data, method=method,
options=options, missing_strategy=missing_strategy,
operating_point=None, operating_kind=None, full=full)
prediction = self.predict_operating_kind( \
input_data,
missing_strategy=missing_strategy,
operating_kind=operating_kind)
return prediction
if len(self.models_splits) > 1:
# If there's more than one chunk of models, they must be
# sequentially used to generate the votes for the prediction
votes = MultiVote([], boosting_offsets=self.boosting_offsets)
for models_split in self.models_splits:
models = self._get_models(models_split)
multi_model = MultiModel(models,
api=self.api,
fields=self.fields)
votes_split = multi_model._generate_votes(
input_data,
missing_strategy=missing_strategy,
unused_fields=unused_fields)
if median:
for prediction in votes_split.predictions:
prediction['prediction'] = prediction['median']
votes.extend(votes_split.predictions)
else:
# When only one group of models is found you use the
# corresponding multimodel to predict
votes_split = self.multi_model._generate_votes(
input_data,
missing_strategy=missing_strategy,
unused_fields=unused_fields)
votes = MultiVote(votes_split.predictions,
boosting_offsets=self.boosting_offsets)
if median:
for prediction in votes.predictions:
prediction['prediction'] = prediction['median']
if self.boosting is not None and not self.regression:
categories = [ \
d[0] for d in
self.fields[self.objective_id]["summary"]["categories"]]
options = {"categories": categories}
result = votes.combine(method=method, options=options, full=full)
if full:
unused_fields = set(input_data.keys())
for prediction in votes.predictions:
unused_fields = unused_fields.intersection( \
set(prediction.get("unused_fields", [])))
if not isinstance(result, dict):
result = {"prediction": result}
result['unused_fields'] = list(unused_fields)
return result